The success of biologic drug discovery and development projects is critically dependent on discovery of functional as well as developable candidates. It requires feeding biopharma pipelines highly functional candidates with desirable developability profiles. Towards this goal, we are analyzing sequence and structural attributes of 91 variable regions (Fv) from 90 antibody-based biotherapeutics, that have been approved for human use, to construct an intrinsic physicochemical profile for biologic medicines. For the structure-based analyses, we have generated homology-based models for Fv regions of all selected antibodies and subsequently performed 200 ns long molecular dynamics simulation to sample the conformational space of each Fv. In this work, we compare molecular descriptors computed using static, energy-minimized models with those computed using conformational ensembles from the collected trajectories. The in silico molecular descriptors can deviate substantially when the conformational dynamics is considered, suggesting that results obtained from the static homology-based models may have been biased by the templates used. Therefore, conformer ensembles may provide a better description of the solution behaviors of the marketed biotherapeutics. We also show how solution conditions such as pH and ionic strength can also influence the flexibility of the antibodies and their associated intrinsic properties.

Towards building an intrinsic physicochemical profile of marketed biologic medicines: Assessment of in-silico molecular descriptors in static and dynamics models

ACS Fall 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.